eduzhai > Applied Sciences > Engineering >

Global Voxel Transformer Networks for Augmented Microscopy

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 24 pages

Abstract: Advances in deep learning have led to remarkable success in augmentedmicroscopy, enabling us to obtain high-quality microscope images without usingexpensive microscopy hardware and sample preparation techniques. However,current deep learning models for augmented microscopy are mostly U-Net basedneural networks, thus sharing certain drawbacks that limit the performance. Inthis work, we introduce global voxel transformer networks (GVTNets), anadvanced deep learning tool for augmented microscopy that overcomes intrinsiclimitations of the current U-Net based models and achieves improvedperformance. GVTNets are built on global voxel transformer operators (GVTOs),which are able to aggregate global information, as opposed to local operatorslike convolutions. We apply the proposed methods on existing datasets for threedifferent augmented microscopy tasks under various settings. The performance issignificantly and consistently better than previous U-Net based approaches.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×